Summary The identification and characterization of novel ?deafness genes? have been a fruitful approach to gain insights into the molecular mechanisms underlying the development and function of hair cells (HCs). The identification of deafness-causing mutations is also a key step towards developing new therapies. However, the pathological mutations are typically found in genes whose products cannot be manipulated favorably with existing drugs. Our preliminary data suggest that a previously unidentified ?deafness gene? in the DFNA27 locus is an exception. DFNA27 is associated with autosomal-dominant, progressive, non-syndromic hearing loss. One of the genes in this locus encodes REST, a DNA binding protein that represses the transcription of its target genes by recruiting histone deacetylases (HDACs). In non-neuronal cells, REST represses the expression of hundreds of neuronal genes. In postmitotic neurons, REST is itself repressed by at least two mechanisms: a reduction of transcription, and incorporation of an alternative exon that contains a STOP codon (exon-S) into the REST mRNA. The objective of the proposed research is to establish how REST function is affected by DFNA27, how REST is regulated in HCs, which genes it represses, and how responsive it is to drug treatment. Our central hypothesis is that alternative splicing of Rest is critical for the de-repression of a cascade of gene-expression changes in HCs, and that HDAC inhibitors can compensate for the DFNA27-caused defect in the alternative splicing-dependent inactivation of Rest in mouse models. This hypothesis is based on our preliminary studies showing that: 1) in mice heterozygous for the global deletion of exon-S (Rest+/dS), HCs degenerate; 2) in mice heterozygous for the HC-specific deletion of exon-S, HCs also degenerate; 3) Rest+/dS mice do not have any obvious defects other than deafness and impaired balance; 4) DFNA27 subjects harbor a mutation in REST; 5) in cultured cells, the DFNA27 mutation inhibits the alternative splicing-dependent inactivation of REST; 6) HDAC inhibitors prevent the loss of HCs in organ of Corti cultures of Rest+/dS mice; 7) impaired inactivation of Rest leads to reduced expression of the epigenetic regulator Baf53b; and 8) the lack of Baf53b causes hearing loss in mice. We propose 3 specific aims: 1) determine the contributions of alternative splicing and transcriptional regulation to the repression of Rest in HCs versus neurons; 2) determine the effects of HDAC inhibitors on the phenotype of Rest-mutant mice and evaluate the frequency of REST mutations in a diagnostic program for the hearing impaired; and 3) establish the effect of exon-S splicing on the HC transcriptome and establish the function of the Rest-repressed epigenetic regulator Baf53b in HCs. These aims will be achieved using a variety of methods ranging from transcriptome analysis to drug efficacy tests in mice. The proposed studies are significant because defining the effect of DFNA27 on the regulation and function of REST will reveal key features of the transcriptional program of differentiating HCs, and has the potential to lead to the development of a preventative treatment for DFNA27-associated hearing loss.